How Industrial Damage Spreads Through an Economy
What Iran’s 2026 war reveals about production networks, blockade exposure and the priorities of reconstruction
By the IRANation Research Desk
The physical damage caused by war is visible.
Factories are struck. Ports are restricted. Production stops. Trade falls.
The wider economic damage is harder to see.
A petrochemical facility may be disrupted in one part of the country, while the resulting shortage reaches plastics, pharmaceuticals, agriculture, construction and machinery elsewhere. A steel complex can lose capacity, but the consequences do not remain inside the steel industry. They move through the firms that depend on its output.
This distinction is at the center of a new study by Mehran Behnia of the University of California, Riverside, and Mahboubeh Davoudi of Sharif University of Technology.
Using the Central Bank of Iran’s official 74-activity input-output table, together with customs, employment and household-consumption data, the authors examine how industrial damage and the U.S. naval blockade during Iran’s 2026 war propagated through the Iranian economy.
Their question is narrower than estimating the total cost of the war.
It is also more structural:
How far can a shock to a small number of strategically positioned industries spread through an economy?
The answer matters not only for measuring wartime damage, but for deciding what Iran should repair first.
The study measures structural exposure, not realized economic loss
The headline figures are large.
Under the study’s benchmark assumptions, the combined war and blockade episode is associated with:
- 15.4% of gross output exposed
- 10.7% of value added exposed
- approximately 1.69 million job-equivalents
- about 14.5% in output-weighted cost pressure
- and roughly $37 billion in compensating-import requirements
But these figures require an important qualification.
They are not estimates that Iran’s GDP actually fell by 10.7 percent. The 1.69 million figure is not a count of observed layoffs. And the 14.5 percent cost-pressure estimate is not a forecast of consumer-price inflation.
The model holds the production structure fixed and asks how much economic activity is connected to the sectors and trade flows that were disrupted.
It does not model all the adjustments that occur after a shock: firms finding substitute inputs, inventories being released, trade being rerouted, production recovering, prices changing behavior, or the government reallocating scarce resources.
The figures are therefore best understood as measures of structural exposure before adaptation.
That distinction is essential.
The value of the study is not that it claims to know the final economic cost of the war before complete national accounts exist. It is that it identifies where economic vulnerability sits inside Iran’s production structure.
Two directly damaged sectors generated a much larger economic shock
The benchmark war scenario applies direct productive-capacity shocks to only two activities:
chemicals and chemical products, and basic metals.
For chemicals, the benchmark treats 55 percent of capacity as unavailable. For basic metals, the corresponding figure is 21.2 percent.
No other activity receives a direct war shock in the baseline model.
Yet the direct effect of those two shocks, equivalent to about 6.5 percent of gross output, grows to approximately 13.8 percent once the effects are propagated through the production network.
The indirect component alone is about 7.3 percentage points.
In other words, the modeled indirect exposure is larger than the initial direct shock.
That is the central quantitative finding of the study.
The reason is not simply that petrochemicals and steel are large industries.
It is that they occupy important positions in the production network.
An economy is not a collection of independent sectors. Industries buy from one another. One industry’s output becomes another industry’s input. Damage to a highly connected supplier can therefore affect production far beyond the facility that was physically struck.
The location of the bomb and the location of the economic effect are not necessarily the same.
Chemicals and metals matter because the rest of the economy uses them
The chemicals sector illustrates the mechanism clearly.
Iran’s petrochemical capacity is heavily concentrated around Assaluyeh and Mahshahr. According to the evidence used by the study, an important part of the disruption came not from the destruction of every individual petrochemical plant, but from damage to shared utilities supplying electricity, steam, water and industrial gases.
That distinction matters.
A utility bottleneck can disable multiple production units even when those units themselves remain physically intact.
Once chemical output falls, the shortage does not remain inside the petrochemical sector. It moves toward industries that use chemical products as inputs.
The study identifies significant exposure in sectors including:
- rubber and plastics
- pharmaceuticals
- textiles
- fabricated metal products
- paper
- electrical equipment
- agriculture
- construction
- machinery
Basic metals create a similar network effect.
Steel is not consumed only as a final product. It enters construction, machinery, transport equipment and investment goods. A reduction in steel capacity therefore restricts activities much further downstream.
The results make this dependence visible.
Under the combined benchmark, chemicals show an 80 percent within-activity output exposure and basic metals 34.9 percent. But several sectors that received no direct war shock at all also register large modeled declines through inter-industry linkages.
Other petroleum products reach 31.8 percent, rubber and plastics 30.7 percent, textiles 23.6 percent and pharmaceuticals 22.5 percent.
The economic significance of an industrial target therefore cannot be inferred from its own contribution to GDP alone.
What matters is also what depends on it.
The war channel was much larger than the blockade channel
The study also separates physical industrial disruption from the U.S. naval blockade rather than treating them as one economic shock.
This produces another important result.
Of the combined 15.4 percent gross-output exposure, approximately:
- 13.8 percentage points come from war-related industrial capacity losses
- 1.0 percentage point comes from blocked exports
- 0.6 percentage points come from disrupted imported intermediate inputs
At first glance, this may appear surprising given Iran’s dependence on maritime trade and the strategic importance of the Persian Gulf and Strait of Hormuz.
But the comparison has to be interpreted carefully.
The benchmark blockade is modeled over an effective enforcement period of only 62 days, roughly 17 percent of a year. The war-related loss of installed productive capacity is not scaled in the same way.
A longer blockade would therefore increase its contribution.
Still, within the episode studied, the larger economic vulnerability came from the loss of productive capacity inside Iran rather than from the temporary interruption of external trade alone.
The two trade channels also work differently.
Blocked exports remove foreign demand. Their effect moves backward toward Iranian firms that supplied exporting industries.
Blocked imported intermediate goods create a supply shortage. Their effect moves forward toward domestic industries that require those inputs to continue producing.
The distinction matters because exports that cannot leave the country do not automatically disappear. Some may be stored, redirected or sold domestically.
An imported component that a domestic factory cannot obtain presents a different problem: production itself may become impossible.
Employment exposure spreads beyond the factories that were hit
The difference between physical damage and economic incidence becomes even clearer in employment.
Value-added exposure is highly concentrated in chemicals and basic metals. Together, the two activities account for about 60 percent of the economy-wide value-added exposure generated by the modeled war shock.
Employment exposure is much more dispersed.
The reason is structural.
Chemicals and basic metals are relatively capital-intensive. Many of the sectors depending on them employ far more workers for each unit of output.
As production constraints spread through the network, employment exposure therefore shifts toward labor-intensive activities.
The study finds substantial shares of employment exposure in agriculture, construction and textiles, even though these sectors are not the original military targets.
Crops and agriculture account for roughly 13 percent of war-related employment exposure. Private and public construction together account for around 19 percent, and textiles approximately 9 percent.
This has a direct policy implication.
Wartime employment support cannot be allocated only by looking at where factories were physically damaged.
A worker hundreds of kilometers from a struck industrial complex may still face economic disruption because the factory where that worker is employed can no longer obtain the inputs it needs.
The larger burden falls on investment, not current consumption
One of the study’s most distinctive findings concerns where the disruption appears in final demand.
Under the combined benchmark:
private-consumption exposure is approximately 6.2 percent.
Investment exposure is approximately 9.3 percent.
The same pattern exists in the war channel alone. Investment exposure reaches 7.3 percent, compared with 4.7 percent for consumption.
This is not accidental.
Chemicals, basic metals and their downstream industries are closely connected to fabricated metals, machinery, electrical equipment, construction materials and other components of capital formation.
Household consumption is more heavily concentrated in food, services, energy and lighter manufactures.
The result is an important asymmetry.
Part of the economic burden of war is visible in present consumption.
Another part appears in future productive capacity.
If investment-linked supply chains remain constrained, an economy can continue experiencing the consequences of war even after the immediate military shock has passed.
Factories may reopen, but machinery may remain scarce.
Infrastructure may need reconstruction, but metals and electrical equipment may remain constrained.
Businesses may want to invest, but the industrial inputs required for capital formation may not yet be available.
The effect of war therefore does not end when immediate consumption begins to recover.
It can continue through slower investment and reconstruction.
The foreign-exchange problem is also a production problem
The study translates the domestic-availability gap into another policy-relevant measure.
Under the benchmark assumptions, replacing disrupted availability through additional imports would require approximately $37 billion in foreign exchange.
The majority of this requirement comes from replacing production lost through the war-related industrial shock. Directly blocked imports account for about $3.55 billion.
Again, the number should not be read as a forecast of the amount Iran will actually spend.
It measures the scale of the availability gap if lost domestic production and blocked goods were replaced through imports under the model’s fixed structure.
The study then asks the opposite question.
What if Iran attempted to replace blocked imported intermediate goods through additional domestic production rather than foreign purchases?
The required domestic gross-output expansion would be about 1.6 percent, equivalent to approximately $7.6 billion in additional domestic gross output under the model.
These are two different pressures.
One is a foreign-exchange requirement.
The other is a domestic productive-capacity requirement.
Confusing them would obscure the actual policy choice.
Importing replacement goods requires scarce foreign currency.
Producing substitutes domestically requires installed capacity, technology, complementary inputs and time.
Economic resilience therefore cannot be reduced simply to either “import more” or “produce everything domestically.”
The constraint depends on the specific input and on where it sits in the production network.
The results remain significant under different assumptions
Any real-time study of an ongoing war faces uncertainty.
The exact capacity lost in chemicals and basic metals cannot be observed with perfect precision, and the effectiveness and duration of a blockade are also uncertain.
The authors therefore vary the main parameters.
In their joint low-exposure scenario, the combined gross-output exposure falls to approximately 10.85 percent.
In the high-exposure scenario, it rises to roughly 20.5 percent.
Employment exposure ranges from about 1.23 million to 2.28 million job-equivalents across those deterministic bounds.
These are not statistical confidence intervals, and the authors do not present them as such.
But the sensitivity exercise shows something important.
The central conclusion does not depend on one exact parameter choice.
Even under the lower-bound scenario, roughly one-tenth of gross output remains structurally exposed.
The study also performs a separate two-directional check to account for the fact that chemicals and basic metals are not only suppliers to the rest of the economy, but also major buyers from other sectors.
That alternative exercise produces a gross-output exposure of roughly 14.8 percent from the two sectors, close to the principal estimates.
The precise magnitude remains uncertain.
The network effect does not disappear.
Reconstruction should follow the production network
This is where the paper becomes particularly relevant for Iranian policy.
If economic damage propagates through networks, reconstruction cannot be prioritized only by visible physical destruction.
The first priority identified by the study is restoring the directly damaged productive base, especially shared utilities serving the petrochemical clusters in Assaluyeh and Mahshahr, and critical stages of the steel chain around Mobarakeh and Khouzestan.
The logic is straightforward.
Restoring one central input can release constraints across many downstream activities.
The second priority is to identify industries suffering because of those bottlenecks.
Rubber and plastics, pharmaceuticals, textiles, fabricated metals and electrical equipment may not have been directly attacked, but their recovery depends on access to chemicals and metals.
They may therefore require priority access to scarce inputs while the central nodes are being restored.
The third priority follows from the investment result.
Reconstruction itself depends on machinery, fabricated metals, electrical equipment and construction materials.
If those supply chains remain constrained, rebuilding damaged capacity becomes slower precisely because the industries required for rebuilding are themselves exposed.
The fourth priority concerns employment.
Support should follow where job exposure appears, not merely where physical destruction occurred.
And the fifth concerns foreign exchange.
When foreign currency is scarce, imports should be prioritized toward intermediate goods, spare parts and repair equipment capable of removing production bottlenecks, rather than treating every lost good as economically equivalent.
The paper also points to the geography of trade.
Customs offices and zones exposed to the blockade along the Persian Gulf and Gulf of Oman account for roughly half of Iran’s goods trade value in the data used by the study.
Reopening the Strait of Hormuz therefore does not, by itself, restore trade immediately.
Ports, customs clearance, shipping coordination, insurance, cargo backlogs and prioritization still determine how quickly the production network reconnects.
Conclusion: economic resilience is a network problem
The main lesson of the study is not simply that war creates large economic costs.
That is already obvious.
The more important finding is that the scale and location of economic exposure cannot be inferred from the scale and location of physical damage alone.
In the benchmark examined by Behnia and Davoudi, direct capacity losses in only two industrial activities propagate through much of the Iranian production network.
The initial 6.5 percent direct gross-output shock becomes 13.8 percent once downstream dependence is taken into account.
Employment exposure appears in sectors far from the original targets.
Investment is affected more heavily than current consumption.
And the resources needed for recovery depend on restoring the parts of the network that other industries cannot function without.
For Iran, this changes how economic security should be understood.
Protecting every facility equally is neither possible nor economically rational.
The more important task is identifying which utilities, industrial inputs, transport routes, ports and production stages have the greatest capacity to transmit disruption through the rest of the economy.
The same principle applies to reconstruction.
Visible destruction tells policymakers where the war hit.
Production linkages help determine what must be restored first.
In a prolonged conflict, that distinction can shape not only the immediate economic cost, but the speed at which an economy is able to recover after the fighting stops.
Sources
Primary source
Mehran Behnia and Mahboubeh Davoudi, How Far Can Targeted Industrial Damage and a Naval Blockade Spread Through an Economy? A Real-Time Input-Output Assessment of Iran’s 2026 War, 2026.
The quantitative estimates and policy findings discussed in this article are drawn from that study.
Principal data and evidence sources used in the study
- Central Bank of the Islamic Republic of Iran, Input-Output Table of the Iranian Economy for 1400 (2021/22).
- Central Bank of the Islamic Republic of Iran, National Accounts, 1403.
- Islamic Republic of Iran Customs Administration, detailed export and import records for the first ten months of 1404.
- Statistical Centre of Iran, Labour Force Survey data.
- Aban Capital Market Services Group, Assessment of the Effects of War-Related Damage on the Country’s Production Chain, spring 2026.
- Associated Press and Reuters reporting used by the authors to establish the war and blockade timeline.
- Argus Media reporting used in the study’s assessment of petrochemical and oil-sector disruption.
- World Steel Association data used to establish the scale and position of Iran’s steel industry.
- Acemoglu, Carvalho, Ozdaglar and Tahbaz-Salehi, “The Network Origins of Aggregate Fluctuations,” Econometrica.
- Carvalho, “From Micro to Macro via Production Networks,” Journal of Economic Perspectives.